Subscribe
Share
Share
Embed
A voice cuts through the track and asks, "What if I fall?" The answer follows: "Oh but my darling, what if you fly?" Erin Hanson wrote those lines at 18. The math underneath them is real, and Sam Bankman-Fried skipped the part that keeps a miss from ending you.
Show Notes
A voice cuts through the track and asks, "What if I fall?" The answer follows: "Oh but my darling, what if you fly?" Erin Hanson wrote those lines at 18. The math underneath them is real, and Sam Bankman-Fried skipped the part that keeps a miss from ending you.
Featuring: Ed Thorp, Erin Hanson, Sam Bankman-Fried
Brought to you by:
Trends.vc: Discover new markets and opportunities.
HeadsUp: Know when your competitors move. Before your customers do.
Multi: Turn any content into a YouTube video.
Where to find Dru:
I write an essay daily. Get them free
Follow me on X
Watch my video essays on YouTube
Follow Dru's Notes on Apple Podcasts or Spotify
Subscribe to Dru's Notes for a new episode every day.
Creators and Guests
What is Dru's Notes?
One mental model each day in a few minutes. Habits, business, building products and thinking clearly. From Dru Riley, founder of Trends.vc.
"And you ask, 'What if I fall?'
Oh but my darling,
What if you fly?"
Erin Hanson
A voice cuts through the track and asks, "What if I fall?" The answer follows: "Oh but my darling, what if you fly?"
Producers keep folding those lines into the dance music I listen to.
The words come from a poem. Erin Hanson wrote it at 18 and posted it to her blog, The Poetic Underground.
The lines traveled far enough that people credit them to Peter Pan.
"What if I fall" weighs the downside. "What if you fly" weighs the upside.
Say a bet costs you $1 when it misses, pays you $10 when it hits and hits 1 time in 3. Run it 30 times and you land around $80 ahead, even though most single bets lose. Gamblers call that a positive expected value bet: the odds pay you if you keep playing.
Odds like that are reason enough to bet before you're certain.
You control whether you take the bet. The result is out of your hands.
Sam Bankman-Fried preached that math in interview after interview: positive expected value, take the bet. He ran FTX, a crypto exchange once valued at $32 billion.
An economist once asked him about a game: 51% it doubles the world, 49% it wipes the world out, played double or nothing.
He answered that maybe you play your way into "an enormously valuable existence."
Gamblers size bets with the Kelly Criterion, a formula that keeps a run of losses from wiping them out.
Bet above Kelly and one miss can end you. Bet below Kelly and your wins stay too small to matter.
He tweeted that he'd bet bigger than the formula recommends.
In November 2022 FTX collapsed.
Customer deposits went down with it and a judge gave him 25 years.
The $80 arrives only if you survive all 30 bets. A 50% loss takes a 100% gain to climb back. Lose it all and the game ends.
Size each bet so a miss can't finish you. Pass on games where a single turn can. When the odds pay, go.
What idea could you test at a survivable size?
(Via position-social → products/multi/.tweet-swipe-cache/what-if-you-fly-brief.md, 2026-07-25. The brief was missing at publish: briefkey was seeded into the first draft commit before any research ran, so a key-only check read as "already done." Step 0a's gate caught it.)
Applied (Dru picked 2 of 3):
1. Name the Kelly Criterion. The post had "a formula that keeps a run of losses from wiping them out" and never named it. Kelly criterion was 1 of only 3 query lanes returning dense on-topic signal at a 250-fave floor (cheat sheets bm1020, Ed Thorp's paper bm918, fractional-Kelly bm391).
2. Two-sided error. The most-bookmarked on-thesis winner (bm2752) closes on "Bet too much and the law of large numbers ruins you. Bet too little and you leave returns on the table." The winning frame is a sweet spot with an error on EACH side, and the post carried only the ruin side. Both levers landed in one edit, since the Kelly paragraph is where the two-sided error is most accurate. Blind gates on the 4 changed sentences: audit-plain-language 4 PLAIN, audit-openers 4 DELIVER, audit-confusion 4 CLEAR.
Declined: sharpening "Pass on games where a single turn can" toward the 726K-view "ruin comes from a single day" framing. The existing line already says it.
Validated (do not churn): the two-sided thesis is genuine white space (roughly 20 winners split cleanly into take-the-bet OR avoid-ruin, none in the middle); the closing line is already the portable/screenshottable shape the data rewards; small-dollar arithmetic beats abstraction; no emoji and no hashtags is correct for this topic.
Query-design warning for future runs: "Sam Bankman-Fried", "asymmetric bet", "what if you fly", "stay in the game" and "can afford to lose" all returned off-topic noise. This conversation lives under quant/trading vocabulary and Taleb's name on X, never under founder-inspiration vocabulary. Details in the brief.